The Utilities menu in the Data Manager contains a number of functions used infrequently to check or influence the way information is stored in Enterprise 6.
You must back up your data before using any of these Utilities, as many of them are not reversible, and many involve long processes which, if interrupted unexpectedly, could leave your data in an unusable state.
Note that there, where a list of duplicates is presented, there is no need to act immediately. One can always save the list as a set for future editing, or search in the contacts’ More Choices for Duplicate State of “D”, and then use the ‘Go to Duplicate’ function to assess the potential deletions.
This function should only be used if it is thought that some of the company Codes have become disorganised in some way, possibly by user error or through the import of companies with codes generated by another system.
Once you have determined, using the usual search methods, which companies are to be recoded, Enterprise 6 will then recode them in the usual format. This consists of the initial letters of the company Name or, if a private individual, the surname of the first contact, followed by the Office Code (set on the User Details page of Preferences) and a unique number. The number of letters taken from the company Name or Surname is determined on the Core Data page of Preferences.
Any reference elsewhere in the database to the selected Companies (for example, in the Diary, the Orders or the Invoices files) are also recoded to reflect the new codes.
This function is as described above (under Recode Companies), except that it recodes the contact Code of selected contacts. If the contact has been used in any other files (for example the Diary), references there to that contact are also altered.
This procedure should be run after data import, whether this occurred from disk or by Remote Access. The latter would include the use of the web for data entry, if the system's normals checks and balances have been by-passed.
It checks for possible duplicate companies either in the entire database or within a selection of companies, which it then lists, giving you the opportunity to decide if and how to delete them.
First, you are asked to make a selection of Companies from which the duplicates are to be removed. This selection could consist of every company in the database. On accepting the list, a dialogue box appears.The Method check boxes determine which fields will be used when making the comparison between two (or more) possible duplicate companies, while the Number of Characters column refers to the length of text used for the comparison. For example, if you have the Town check box on, with a two character setting, Enterprise 6 will put all companies in London and Loughborough forward as possible duplicates as their towns begin with the same two letters. If the character setting is three, they will not be considered as possible duplicates.
The final option, Consider blanks as matches, is a further control to prevent unwanted deduplication being carried out by mistake. If turned on, two Companies, for example, with a similar name and no Postcode could be considered duplicates. As more Method check boxes are switched on, the more definitive, but also the more time consuming the process will be.
Once the checks have been completed, the possible duplicates are listed. You must now decide how to deal with them. The options available are shown in the dialogue box, which comes up after you have pressed the [Delete] button in the palette. If you choose one of the first three options, highlight only those items in the list to be deleted and then press the [Delete] button in the palette.
The first option will delete only the company record(s) selected.
The second option will delete the company record(s) selected and related contacts, Diary Items, Orders, etc.
The third option will delete the company record(s) selected and transfer this related information to the company Code entered to the dialogue box. Note that as this is a dialogue box, the wildcard (@) cannot be used to bring up a selection list, so take care to type the company Code correctly. Note that all Companies selected to be deleted will have their related data transferred to this one company - Enterprise 6 will not delete them one by one and repeatedly ask for a new company to which it is to transfer information.
The fourth option will probably be the most common solution. This enables you to delete one of the duplicate companies and leave the other in the system. If you do this, you will probably want to transfer all contacts, diary items, orders, invoices etc to the company remaining in the database. If you want to follow this method, highlight all the duplicates to be deleted and the Companies to which their data is to be transferred before pressing the [Delete] button in the palette. After selecting the final option, Enterprise 6 will then ask you to re-sort and will perform the deletions and information transfers by leaving the first company in the sorted list in the database and deleting the other(s). This option is the most intelligent of the four. If, for example, you find you have two Apple Pie Trading Companies and three Fruit Production Companies, you can highlight all five at once and press [Delete]. Enterprise 6 will delete one Apple Pie Trading company and two Fruit Production Companies.
This function is similar to Deduplicate Companies and you may need to run it after that has been completed if you transferred related information to another company. This process often results in duplicate contacts being created.
On selecting this function, you are first asked, ‘Do you want to find all contacts that have not been checked, or to select manually?’. If you choose to [Select] you will be shown the normal contacts Quick Search window, from where you can choose the More Choices window if necessary, whereas if you choose [Not Checked] Enterprise 6 will search for all contact records with a blank Duplicate State field.
Once a selection of contacts has been made, Enterprise 6 will look at each contact and check for duplicates across the whole database rather than merely within the selection. The ones that match could be other new entries or they could be older contact records that have previously been checked (ie had their Duplicate State set to “C”), and were at that time found not to be duplicates. Whether you decided to [Select] or to work with all contacts [Not Checked], the contacts found will be listed, so that you can alter the selection in the normal way using the Buttons Palette if necessary.
After confirming the list by clicking [Ã], a dialogue box appears.It asks you which fields and how many characters of those fields you want to employ for deduplication. The Method check boxes determine which fields will be used when making the comparison between two (or more) possible duplicate contacts, while the Number of Characters column refers to the length of text used for the comparison. For example, if you have the Town check box on, with a two character setting, Enterprise 6 will put all contacts in London and Loughborough forward as possible duplicates as their towns begin with the same two letters. If the character setting is three, they will not be considered as possible duplicates.
The final option, Consider blanks as matches, is a further control to prevent unwanted deduplication being carried out by mistake. If turned on, two contacts, for example, with a similar name and no Postcode could be considered duplicates. As more Method check boxes are switched on, the more definitive, but also the more time consuming the process will be.
There is a check box below the Method options ‘Check Multi-Companies’, which is switched off by default. When it is off, only the Town/Postcode etc of the first company for each contact is considered during the deduplication searches. When it is on, all the contact’s Companies are considered, and if any of them conflict, the contact will be considered a duplicate. The latter option would therefore be preferable in most cases, but will further slow the checking process. The ‘Check Multi-Companies’ option cannot be used in combination with the Address Line 1 check.
Also at the base of this window is a check box ‘Overwrite existing Duplicate States’, which can optionally be turned on if you are re-checking contact records that have previously been checked. The default is not to overwrite these States. When you click [OK], Enterprise 6 will record a Duplicate State of “C” for each contact that has been checked and found not to be a duplicate, and “D” for each one found to be a duplicate.
Note that if two or more contacts are in the list to be checked (ie included in the initial selection list shown on screen) and are found to be a duplicate against each other, then they will each be recorded as “D”, whereas if one being checked is found to be a duplicate of one that had previously received a ‘C’ (and therefore was not in the list to be checked) then the latter will not be marked as a “D” (even if you are using the Overwrite option). In other words the Duplicate State field will only ever be changed for those records that were listed to screen before commencing. For all the duplicate records (ie both those in the list and those previously checked), a note is appended to Comments field as follows: ‘1/1/95 Duplicate of L/1111’ - ie the date of the checking, and then the contact Code of the contact record that matches the one on screen.
Once this function has been completed, you can use either this Comments field text manually to note and find the conflicting contact record, or the contact function ‘Go to Duplicate’ to carry out the same task automatically. Full details of this function can be found in the Companies and contacts reference chapter.
Before any of the deduplication processing actually takes place you are given one more message: ‘NB: Any records that are locked by other users will not be updated by this process, so will appear again as 'unchecked'. All records that are checked will become locked to other users.’ Essentially this is an advisory warning which indicates that it is wise, but not imperative, that other users are either not using the system at all, or at least are not editing the contacts file.
Once the deduplication processing has completed, you’ll be shown a list of the potential duplicates. This list will include both records that you selected for checking and those that they have been matched against. Here you would double-click, review the records, use the ‘Go to Duplicate’ Function, and decide which ones to delete. When decided, you press the [Delete] palette button and can use the deletion option within the ‘Transfer’ section: transfer contact information ‘to the last specified Duplicate’. This option overrides any text you type into the box above, and means that, for each contact selected for deletion, Enterprise 6 looks up the Duplicate contact Code last specified in (ie at the bottom of) its Comments field, and transfers other details to this contact. If the contact specified in the Comments field does not exist, or if the Comments contains no suitable data you’ll get a message to that effect and no deletion will occur. If the contact that is pointed to does not exist because it has just been deleted by the very same deletion process, then again no further deletion will take place, but there will be no message. This means that you can highlight and [Delete] a pair of contact records that point to one another and have only one deleted. That is not to say that you shouldn’t highlight the exact one to delete - that will always be better, especially as otherwise the first one encountered will always be the one to be deleted.
Another check box in this window is ‘Delete contacts’ Companies if uniquely theirs’. This is applicable to only the ‘transfer related info’ option. It means that the result of the contact deletion can also be a company deletion. The transfer process without this option will perpetuate company duplicates, carrying them forward to the newly united contact - and they then have to be removed using ‘Deduplicate Companies’. When this option is turned on, for each company that is linked to a contact being deleted, the system assesses whether the company or address is also used by any other contact(s). If it is not, ie if it ‘uniquely’ belongs to the contact being deleted, that company record will be deleted, and its related data will also be transferred. The code to which the data will be transferred is the first company Code of the contact to which the contact data is being transferred. When the Deduplication process is finally over (either it has found none or you have put away the listing screen), you are asked ‘Do you want to save any changes to the Duplicate State and Comments fields, and to confirm any deletions made?’. If you say [No] to this question, all of the changes that were visible while you were editing the list and all of the deletions (if any) that you have made will be discarded. You would have reverted to exactly the state the data was in before you started the process. This is obviously a useful escape method, though it should only be used if absolutely necessary due to the time-wasting it implies.
