Automate really effective direct messages on twitter

Recommendations, examples and alternatives

All social networks harbor the occasional bad practice and Twitter is no exception. Among them, automating the sending of messages to new followers could well be considered one of the most dangerous of being annoying to the user and penalized by Twitter, but before we explain what this technique consists of.


What is an automatic direct message

An automatic direct message (or auto DM) is a message that the user instantly receives by following a Twitter account. Initially these were sent by Twitter to the email of the account owner, a feature that is now disabled by default in all accounts.


If you have not had to experience it in the first place, you may have heard comments about it. In our case, during these last 10 years we have been able to see an infinite number of ideas and applications where this technique is automated, from the most basic to the most innovative ones, generating positive and negative reactions in users.

This experience has served to draw lessons, conclusions, identify good practices and, not least, know closely the consequences in regards to users and Twitter in particular. Information that we have collected and ordered to help you learn everything you need to know about it and with which you will make better decisions when you consider applying it in your particular case.

Hi!
Thanks for follow me. You can also follow me on this other social network

Yesterday at 4:51

Of course we are going to show you step by step how you can automate the sending of direct messages, but also what are the latest techniques and trends, all the options you have, in addition to sharing a multitude of automations that we have extracted from real cases and that we have left preconfigured so you can use them.


One of the techniques you are going to learn is the use of filters. In this formula we share, we use them to send DMs only to followers whose accounts are less than X months old.

Users of recently created accounts may become more interested in receiving a new MD.


Does Twitter allow the automation of DMs?

Let’s start by clearing one of the most common questions. The automation of DMs (Direct Message) is allowed , as stated by Twitter in its policy of rules for automation, however, the express consent of the user is required and conditions are established to comply for its use.

Twitter is one of the few social networks that from its origin offers a multitude of ways to access and operate with data from its platform. Despite the latest movements of the social network towards a more protected ecosystem and more limited access to its data, it continues today to be one of the platforms with the greatest potential for automation.

Paradoxically, from this opening, bad practices have also emerged, where the biggest affected end up being the users of the platform, who are exposed to the publication of unwanted messages and interactions and against which they have few tools with which to defend themselves.

Thanks for follow me!
Follow me also on this, this and this other social network

Yesterday at 4:51

To protect both users and the platform itself from practices that are aggressive or could compromise the service, Twitter established in late 2017 a policy of specific rules and recommendations for automation. This is what it says about direct messages:

You may not send unsolicited Direct Messages in a bulk or automated manner, and should be thoughtful about the frequency with which you contact users via Direct Message. You may send automated Direct Messages to users so long as:

  • in advance of sending the Direct Message, the recipient(s) have requested or have clearly indicated an intent on Twitter to be contacted by you via Direct Message, for example by sending you a Direct Message; and
  • you provide a clear and easy way for such users to opt-out of receiving automated Direct Messages, and promptly honor all such opt-out requests.

The fact that a user is technically able to receive a Direct Message from you (e.g. because the user follows you, has enabled the ability to receive Direct Messages from any account, or because the user is in a pre-existing Direct Message conversation with you) does not necessarily mean they have requested or expect to receive automated Direct Messages from you.

Source: Twitter Automation rules.

The automation of DMs is allowed, however this requires the express consent of the user; Consent that is not obtained due to the fact that a user decides to follow you on Twitter, which initially invalidates any implementation of this technique and explains why Twitter occasionally reserves the right to allow or deny the use of this automation at its discretion.

Finally, the consequences of violating this rule vary and range from disabling the sending of DMs to the temporary or complete suspension of both the Twitter account that operates and the application that performs the automation.

We will talk about the consequences that this can cause to the brand or to the relationship with users.


When is advised to do it?

Taking into account and known by all the low predisposition of users to receive automatic messages and the opinion of Twitter in this regard, it is necessary to take precautions as we will see below, but also consider the following aspects as a starting point.

  • Make sure you have found an original implementation whose added value also satisfies a considerable percentage of your users.
  • Test your strategy with a small group and confirm that the answer is satisfactory.
  • Evaluate the possibility of responding manually and the time it would take you according to the volume of new followers you manage. Does it justify automation?
  • Make sure that you are going to have control over the process and you can interfere if something unforeseen occurs.
  • Also make sure that you will be able to progressively scale the automation process as you validate the results.

Validating any strategy is easy if you use filters to segment.

You can test it with a small number of users, with a formula like this in which it will only apply to those with less than 50 followers in their accounts.


Step-by-step automation

The process to have your own DM autoresponder is simple and it will only take you a few minutes to execute it if you follow these steps, or even less if you use any of the predesigned formulas that we share throughout this article.

  1. Access Twitter and create your developer account if you don’t have one. Here is the complete process to follow to create a developer account.
    Twitter developer page
  2. Activate the option that will allow you to read and send direct messages in an automated way by accessing to https://developer.twitter.com/en/apps and selecting the application you have created using the Details.
  3. Click on the Permissions tab and hit the Edit button.
    Select the "Read, Write and Direct Messages" option and save the changes with the button Save.
    DM read and send permissions
  4. Access Botize with your account.
    Botize login form
  5. In the My tasks section click on Create a new task.
    Create a new Botize task
  6. Once in the task editor, select the Twitter application.
    Automate a trigger with Twitter
  7. If you already have a Twitter account linked to Botize, select it, otherwise press the Connect to Twitter button.
    Connect Botize to Twitter
  8. Enter your access codes that you obtained in the first step.
    In the guide you have all the steps to obtain the keys.
    Connection keys to Twitter
  9. Now select the If a new user follows me, option, so the task will fire with each new follower.
    Twitter trigger If a new user follows me
  10. Press Next and select the Twitter application again to configure the action this time.
    Automate an action with Twitter
  11. Select the option Reply to the user with an automatic message.
    Twitter action Respond to the user with an automatic message
  12. In Message write the one you want to be sent to each new follower.
    Configuration form for direct message response
  13. Click on Save Changes and you’re done.
    Respond by DM every time a user follows me

From this moment on, every hour Botize will check if you have received a new follower, in which case it will send you the message you have programmed and will wait another hour to repeat the same process with the next new follower you may have received. Repeating the process indefinitely

The frequency, the amount of messages and many other aspects are configurable as we will see now.


Predesigned formulas

To make the process even easier you have formulas, which are tasks already predesigned by other Botize users and that you can directly use or customize according to your needs.


7 mistakes you should avoid

It is easy to fall into bad practice even with the goal of helping in mind. It happens frequently and to a greater or lesser degree it can end up affecting the brand and your relationship with users, so the widely-repeated recommendation not to repeat and learn from past mistakes also applies here.

  1. If there is a formula in this case for failure, this is to use the direct message to promote a product, your other social networks or the last blog post. Trying to sell, directly or indirectly at the first contact, is a strategy that we have seen repeatedly fail.
  2. Less obvious but just as dangerous is automating messages in newly created accounts. For some reason the odds in this case of ending up being penalized by Twitter increase.
  3. The same happens if you include links in all the answers or if they always refer to the same domain name. In both cases the chances of receiving a penalty are also increased.

    You will know that you have received a first Twitter notice if you encounter the following error message:
    “Error 226. This requests looks like it might be automated”.
  4. As with URLs, the use of hashtags in all messages or the use of the same hashtags results in penalties.
  5. Automating 100% of the activity of an account, eliminating any manual activity, seems to trigger alarms in Twitter’s algorithms that are in charge of detecting sensitive behaviors of incurring some bad practice.
  6. Not so well known is the problem derived from the speed or frequency at which you schedule the sending of messages. Twitter sets a limit of 1000 direct messages a day, however respecting the limit is not enough.

    Experience tells us that the total number of messages and the speed with which you send them is related to what Twitter considers "aggressive practices" and also increases the possibility of penalization..
  7. Without sufficient data to allow us to affirm it, there seems to be a certain correlation between the percentage of DMs sent for which there has been no response or interaction by users with the probability of resulting in a penalty.

So even if you can, no, definitely don’t use auto-DM to automatically launch promotions to users. Users are especially sensitive to spam and quickly detect what is and what is not a promotional message. We are skilled even differentiating an automatic message from one written and sent individually.

Action Probability of penalty
Automate DMs in newly created accounts high
Frequent use of URLs in messages high
Frequent use of hashtags Medium
Using apps with a negative history Medium
Frequent use of the same URLs Medium
100% automated accounts Medium
Frequent use of the same hashtags Medium
Negative or no response from users Low

Scheduling an immediate auto response can demonstrate the use of automation.

Letting hours or days pass is more in line with the natural behavior of users.

Limiting responses to a schedule is another way to measure work in a way similar to how you would organize yourself.

This other formula is programmed within the schedule that you consider appropriate.


Limits on the use of DMs

In 2015 Twitter expanded the number of characters you can use in a DM message from 140 to 10,000 characters.

Until 2017, the maximum number of daily messages that Twitter allowed you to send went from 250 to 1,000 messages. Feature that was born along with the launch of the first version of the chatbot platform.

These limits are used as a tool to avoid the abuse of both the resources of the social network and the users of the platform. Security measures not exclusively from Twitter but from other social networks such as Instagram, Facebook or YouTube.

Maximum number of DMs allowed 1,000 messages per day
Maximum allowed message length 10,000 characters

Currently if you reach these limits Twitter warns you by means of an error message. As for the daily message limit, once has been exceeded, you can continue with the delivery after the necessary time has elapsed. Every day Twitter restarts the counter and you have the 1000 message quota again.


Best practices

The action of sending a direct message does not hide danger. It is the message and the objective behind it that we must take care of, being self-critical and prioritizing providing value and helping the user with it, a task that, everything is said, hides enormous complexity, but experience seems to show that it is the best recipe to achieve start conversations.

The trigger for the action you expect the user to take does not appear to be sending an automatic message, but rather to start a conversation with him..

This approach, fostering conversation, is more aligned with the philosophy of Twitter and what users of a social network seek or expect.

To achieve this, the actions that are giving the best results are:

  • Take care of the style in the writing of the messages. Reflect your way of writing, keep your essence.check
  • Address the user the same way you do when you write tweets.check
  • Think about the message you are going to write keeping in mind what users want to hear from you. We will see an example.check
  • The surprise factor is not necessarily a positive one.
    Consider communicating in advance on your timeline or in your "bio" how you are making use of direct messages.
    Or in what way you are helping or thanking each new user who gives you follow.check
  • Maintain a balance between automated and manual operations.
    100% automated accounts are more sensitive to result in a penalty.check
  • Automate shipping but don’t do it immediately.
    Immediate evidences automation unnecessarily.check
  • For Twitter, the speed at which you schedule messages to be sent is important.
    Reduce it as much as possible. The longer the time between sending and sending, the less chance you will be penalized.
    Avoid Twitter considering your practice as aggressive.check
  • Start by automating delivery to a few users.
    Correct and refine your message with the feedback or response rate you get.check
  • The value you add to the message works best when it’s in the message itself.
    No URLs to posts, landings, videos, or other web pages. We seek to start conversations.
    Under your message he has a text box to reply to you. Don’t send him somewhere else.
    check

The number of users that an account follows can be an indicator of the volume of messages that it usually handles.

Accounts that follow thousands of users may in some cases find it more common to receive DMs.


Add value in the message

Easy to say, complex to achieve. You need to know your followers, what they have in common, what characterizes them or what qualities they share. If you still cannot answer it with confidence it may mean that it is too soon to schedule the sending of messages with guarantees.

One way to help you respond is by asking yourself why are your followers? What do they expect from you or your account?

If it is still difficult to find answers to these questions, our experience tells us that your account or Twitter profile may not be everything vertical, specialized or single-topic that this type of automation requires to operate. This does not mean that you should change it, but it does mean that there are other more appropriate automations for your case, such as the ones we are going to see now.

The better you know your followers, the better you can find one or more common or highly probable characteristics that you can take advantage of to target them with the autopilot on.


An example of automatic message

A common characteristic to 100% of all your followers, whether you are a brand or not and regardless of the content you offer is that they all use Twitter, at least at the moment they decide to follow you are users of the platform.

How could you help a person knowing only that they have Twitter and that by just following you has shown a certain degree of interest in your content or your profile?

To answer, let’s imagine that you usually talk about web design or programming, although it would be applicable as you will see to other sectors, this example is not valid for everyone.

Probably your followers share or have some relationship with this sector, or they will follow you because they are in the process of making, updating or improving their respective sites and their recommendations or knowledge in the matter is useful to them.

With this in mind, we could think that we could automatically reply with a message similar to this and that we would make little mistake.

Hi!
I have seen your website and I found it very interesting. I have also found improvements that would help you with search engine positioning

Yesterday at 4:51

The attempt is good, however:

  • We would not have visited the user’s website.report_problem
  • We would not have looked for or found improvements to make on your website.report_problem
  • We would not even know if the user has a website.report_problem

Some people may find it strange or uncomfortable even the fact that following your account has caused such dedication or deployment of resources on your part.

To feel natural, the message must add value in its proper measure. Not to mention ethical or not to affirm actions that have not taken place. A more appropriate way could be with this other message.

Hi!
I would suggest some improvements to your website that I think will help you in search engine positioning. In any case, thanks for following me!

Yesterday at 4:51

This other message, equally effective, opens the possibility of initiating multiple conversations and in which the user does not have to feel overwhelmed by the previous work you have done, leaving the reader’s interpretation of the degree of dedication (a list of suggestions with greater or lesser degree of application to the reader’s particular case), even eliminating any possible feeling of debt that the first message might have generated for the user.

Clear, effective and conducive to starting a conversation.

Other examples for whom no possible variation of this case applies:

  • Offers a free consultation of several minutes only to new followers.check
  • Give a product or service prepared for it just by following you.check

As for whether the users you are targeting have a website or not. You can minimize the error by filtering and targeting only those users who have a specified URL in their profile.

This formula uses the same criteria to filter.


Think about how to capture the user’s interest so that he decides to start a conversation with you.
Whatever you want to achieve requires or will be enhanced with this first step.


Applications to automate sending messages

There are applications dedicated to sending DM or that include it as part of other possible automations. The main differences and aspects to consider are:

  • Sending speed.
    Sending in bulk or in short time intervals can result in a penalty from Twitter.
    It is important that you can control the publication / hour ratio. check
  • Control over the final result.
    Some applications may add to the end of your message hashtags or URLs.
    Both elements, in addition to being sensitive to penalties, confuse and demonstrate the use of tools.check
  • Cancel button.
    It seems obvious but make sure you have identified the mechanism to pause or stop automation if things get complicated. check

It is also recommended to be able to use your own private application, something that many users do not know they can do as it is not an option that is usually offered. As a general rule, any automation with Twitter requires connecting your account to the application that will allow you to automate a process.

With this you are giving your consent so that the application can operate, in the same way that it is already doing with other users.When Twitter detects bad practices, it identifies from which application they are carried out and can temporarily or indefinitely block both the account and the application.

This is one of the main reasons why many of the applications that we know in the Twitter ecosystem frequently stop working. The best way to avoid collateral damage as a result of other users’ bad practices is to use your own and exclusive application credentials.

In Botize the automations are carried out from your developer application on Twitter, which means that you make an exclusive use, you do not share the application with other users and also the resources and limits of use established by Twitter for each application, there is no need to distribute them the more accounts.

Another advantage is that in this way you have control over the permissions you are going to give to operate your account and not necessarily accept the permissions that the application considers.

Application It has a free version Alter or modify the original text Use your app, not a generic one
Crowdfire Yes Yes No
SocialOMPH No No No
TuitUtil Yes Yes No
TweetTunnel No No No
Botize Yes No Yes
AppUnfollowers Yes No No
TwitterDMer Yes No No
Twittmate No No No
IFTTT Yes No No
Uniclix No No No
Horl No No No
TweekSocial No No No
TweetManager Yes No No

Alternative to self-DM: Do it publicly

As obvious as it may seem, you can, if you do not already, use your Twitter account to promptly promote the services of your company or blog if it is what you are looking for. Far from being annoying it is something predictable and accepted.

What can make you think that the same message per MD will have a greater impact than its equivalent in tweets when we know that it is not liked by either Twitter or users?


White Hat Alternative: De-automate the sending part

From the process of searching for new followers, composing the reply message and sending it, everything except the last step is perfectly automatable without taking the slightest danger.

If at some point you have had the opportunity to manually run the entire process, probably and to optimize your time, there will have come a time when you considered starting to filter some followers and only go to those that you considered potentially most interesting or willing to interact with you.

You will have verified then that the action of sending the message was everything you should do, the simplest part and that it consumed less time, so it makes sense to think about programming only the first steps.


This other formula identifies potentially more influential new followers with at least 5000 followers in their accounts.

When they find them, in this case it sends you an MD with the details, advising you so that the step of writing to the user takes you done manually.


Non-intrusive alternative: Welcome messages

In 2017, with the rise of the first chatbots, Twitter introduced the possibility of customizing welcome messages, which are displayed when a user accesses your profile in order to send you a DM.

These messages are less intrusive and are being configured as one of the best ways to contextualize a conversation. It also serves as a start message for attendees or chatbots within the Twitter platform.

While Twitter allows you to set a specific welcome message, with Botize you can make it dynamic and easily adapt it to different scenarios.

This formula defines a list of welcome messages that Botize will rotate weekly.

This other formula is designed so that you can personalize the message that will appear from Monday to Friday, for example informing about business hours, along with a different message for weekends, for example inviting the user to send an email to support during the days off.


DM as notification and alert center

The possibilities of automation do not end and the community continually finds new and original ideas, all with the common goal of helping you save and optimize your time.

Some go through directing the DM to specific users, usually followers or colleagues.

This formula is in charge of sending the details of the new appointments of the agenda to a certain user as well as they are approaching, which you may well be yourself and use it as a reminder.

Sending DMs to yourself is a very little exploited feature and on the other hand offers many possibilities.

Send yourself a notification every time you receive an email in your email from a specific account, which contains certain words.

Or send alerts when events occur that otherwise require intensive monitoring.

Save the time you invest in periodically going to other data sources to consult information.

You can bring them to Twitter and receive them by DM if they are private data or do not need to be shared with anyone.

There are many other important but very unappreciated tasks that you can also automate and channel through DM.

Take a look at the integrations that you have available in Botize and you will surely find dozens of ideas.

Others can save us a process that is manually too laborious or complicated.

Inexplicably, downloading the GIFs or videos that we like from a tweet can be quite a challenge.

And of course, life would be too boring if we didn’t always keep a space for videos, or in this case GIFs, of kittens.